centrosome and microtubules behavior in the...

J. SUBMICROSC. CYTOL. PATHOL., 27 (3), 381-389, 1995

Centrosome and microtubules behavior in the cytoplasts

L.A. GORGIDZE and I.A. VOROBJEV A.N. Belozersky Institute of Physico-Chemical Biology, Moscow State University, Moscow, Russia

SUMMARY - After enucleation of PK (pig kidney embryo) cells using cytochalasin D centrioles remained approximately in 80% of cytoplasts. Some cytoplasts retained a single centriole. 14-16 h after enucleation large secondary lysosomes and lipid droplets were evident around the centrosome of many cytoplasts. In part of the cytoplasts replicating centrioles were found 16 h after enucleation. Ouabain treatment (1 mM, 30 min) of the cytoplasts resulted in the appearance of mainly perpendicular orientation of mother centrioles relative to the substrate surface. Reconstruction of microtubule pattern around the centrosome showed that a total of approximately 15 microtubules were attached in the centrosome of normal cells and twice more than that in cytoplasts. Microtubules in the cytoplasts were more resistant to nocodazole induced depolymerization. We suggest that microtubule instability is modulated by regulatory effector that is under direct nucleus control.

KEY WORDS centrosome - centriole - microtubule - cytoplast

INTRODUCTION

The centrosome in mammalian cells has two centrioles: a mother (active) centriole and a daughter (inactive) centriole (Vorobjev and Chentson, 1982; Alvey, 1986). Active centriole carries several pericentriolar satellites from which most microtubules radiate (Vorobjev and Chentsov, 1982). In addition to the two centrioles, the centrosome may contain free foci from which microtubules also radiate (Peterson and Berns, 1980; Vorobjev and Nadezhdina, 1987). Detailed electron microscopic studies show a complex organization of microtubule array around the centrosome (Vorobjev and Nadezhdina, 1987; Alieva et al., 1992), Particularly, in different cultured cells only few long microtubules were fastened there. Others had their free

Mailing address-. Dr. Ivan Vorobjev, A.N. Belozersky Institute of Physico-Chemical - Biology, Moscow State University, 119899 Moscow, Russia.

ends directed towards the centrosome (Alieva et al., 1992). Radial system of microtubules emanating from the centrosome is highly dynamic and undergoes dramatic changes after a variety of treatments (Schliwa et al., 1979; Alieva et al, 1992; Joshi and Baas, 1993). ' Microtubule system and microtubule dynamics in the cy-toplasts was described as being similar to that in normal cells (Brown et al, 1980; Karsenti et al, 1984) and it is assumed that no changes happen in the centrosome-mi-crotubules relationships during enucleation (Kuriyama and Borisy, 1981). But no direct comparison of the cen-trosome structure and microtubule pattern around the centrosome of cells and cytoplasts has been made. In the present study we reinvestigated microtubule system in the cytoplasts, putting emphasis on the microtubules associated with the centrosome. Centrioles in cultured cells usually occupy perinuclear area. They are randomly oriented towards each other (Alvey, 1986; Vorobjev and Nadezhdina, 1987) and to the substrate surface (Albrecht-Buehler and Bushnell, 1979; Vorobjev and Nadezhdina, 1987). But in a number of cases mother (active) centrioles show preferentially perpendicular orientation to the substrate plane. This occurs

Centrosome and microtubules in cytoplasts 381

when mouse fibroblasts are spreading on the coverslip (Gudima et al., 1986), and in PK cells under the effect of 2,4-dinitrophenol, carbonylcyanide-p-trifluoro-methoxy-phenylhydrazone (FCCP) and ouabain (Alieva and Vo-robjev, 1995). Using cytoplasts obtained from tissue culture cells we tested the idea whether the induced change in the angle of inclination of active centrioles relative to substrate plane is an autonomous, independent of the nucleus, characteristic of the centrosome. Studying cytoplasts we found that after enucleation a) the number of microtubules fastened to the centrosome structures increased, b) cytoplasmic microtubules became more resistant to the nocodazole treatment and c) mother centriole was able to orient perpendicularly to the substrate surface.

MATERIALS AND

METHODS Tissue culture

The experiments were carried out on the PK (pig kidney embryo) cells. The cell culture was grown in medium 199, supplemented with 10% bovine serum and antibiotics (streptomycin + penicillin or gentamycin). The cells were subcultured in plastic Petri dishes on coverslips in 5% CO2, at 37° C. Cytoplasts were obtained using cytochalasin D. The final concentration of cytochalasin D in the culture medium was 1 µg/ml; the cells were incubated for 4-6 h. After that the coverslips with cells down were mounted in capron chambers and centrifuged at 10,000 g for 1 h at 37° С in a JS-13 rotor on a J2-21 centrifuge (Beckman Instruments). Coverslips were washed thrice with pure culture medium, each time for 10 min, and then placed into Petri dishes in a thermostat until fixation. Rhodamine 123 staining of cytoplasts was performed as described else-where (Johnson et al., 1980, 1981). The preparations of living cytoplasts were observed and photographed on Opton Photomicroscope-3 (Ger-many) using an RF-3 B/W film (1500 ASA). Ouabain was added to the culture medium at a final concentration of 1 mM, and the specimens were fixed 30 min later.

Electron microscopy

Cells (cytoplasts) grown on the coverslips were fixed with a 2.5% glut-araldehyde (Merck) on phosphate buffer (pH 7.3), postfixed with 1% osmium tetroxide, then stained with uranium acetate, dehydrated and embedded into Epon 812 mixture. For a detailed study of the structure of the cell center the cells (cy-toplasts) were lysed, prior to fixation, for 15 min in the following solu-tion: 25 mM K-Na-phosphate buffer (pH 6.8); 1 mM EGTA; 1 mM MgCl2; 0.5% (w/v) Nonidet NP-40; 4 M glycerol. After lysis specimens were fixed with a 1% glutaraldehyde (Merck) on phosphate buffer (pH 6.8) for 1 h. Further preparation for electron microscopy was as above. Serial ultrathin sections were prepared on an LKB-3 ultramicrotome (LKB, Sweden), serial semithin sections (0.3 цт) were obtained on an Ultrocut ultramicrotome (Reichert-Jung, Austria). The sections were mounted on single slot grids, covered with a formvar film, and then stained with lead citrate according to Reynolds. The ultrathin sections were observed and photographed using electron microscopes HU-11B and HU-12 (Hitachi, Japan) operating at 75 kV. The spatial analysis of microtubule pattern around the centrosome was

382 GORGIDZE L.A. and VOROBJEV IA.

carried out on stereophotos of serial 0.3 µm sections. Stereopairs were photographed at a tilt angle of 10° (magnification 12,OOOX) on an H-700 electron microscope (Hitachi, Japan) operating at 150-175 kV.

Microtubule pattern and centriole orientation analysis

The analyzed series included all the sections containing centrioles, and two additional sections: one above the centrioles and the other below them. Microtubules under reconstruction fell within the area 2 X 3 µm. Analysis of the stereophotos allowed us to superpose the end of each microtubule, running beyond the limits of one section and its continuation on the other section as well as to identify all the microtubules, fastened in the cell center, or having free proximal ends within the reconstructed part of the cytoplast or of the cell. Microtubules were classified according to the following characteristics: the length of a microtubule, location of its proximal end relative to one of MTOC (free focuses of nucleation of microtubules, heads of pericentriolar satellites, walls of cylinders of active and inactive centrioles) and the distance from the proximal end of a microtubule to these MTOC. According to their position and remoteness of their proximal ends from MTOC microtubules were classified under two classes (modified from Alieva et al., 1992): 1) attached microtubules, their proximal ends are offset by no more than 100 nm from the centriole surface (or from the head of a peri-centriolar satellite, or from one of the electron-dense free foci); 2) non-associated microtubules, with their proximal ends offset by more than 100 nm and directed towards one of the microtubule convergence structures. In addition, we took into separate account microtubules running in the vicinity of the centrosome, but not directed to the centrioles,, satellites or free foci. The inclination angle of the centriole cylinder to the substrate plane was determined by the length of projection of centriole microtubules on the section plane (Albrecht-Buehler and Bushnell, 1979). The centrioles having projection length less than 0.03 µm were determined as perpendicular (Alieva et al., 1992). The percentage of perpendicular centrioles obtained was compared with that expected for the randomly distributed centrioles (Vasilyev et al., 1988).

Immunofluorescence

Immunofluorescent staining of the microtubules was performed accord-ing to the indirect method as described elsewhere (Bershadsky and Gelfand, 1981). Cells grown on the coverslips were lysed in microtubule stabilizing medium, then fixed with glutaraldehyde (0.5-1%) on phos-phate-buffered saline (PBS) processed with borohydride and stained with primary anti-a-tubulin monoclonal antibodies (Sigma), then with anti-mouse FITC-conjugated IgG (Sigma).

RESULTS

The method of enucleation used gave up to 90% of cy-toplasts. Rhodamine 123 staining revealed in cytoplasts a network of mitochondria uniformly filling the cytoplasm (Fig. 1 a,b). However bright fluorescence of mitochondria was not observed in all the cytoplasts cultured overnight. In cytoplasts of the polygonal shape the centrosome was located approximately in their geometrical center. In cy-toplasts of the spindle or branched shape the centrosome was often located in one of the cytoplasmic lobes.

FIGURE 1a,b The fluorescence of mitochondria in the cytoplasts stained with Rhodamine 123. (a) Rhodamine fluorescence; (b) phase contrast. Bar = 10 µm. FIGURE 2a-d Antitubulin staining of cells and cytoplasts. (a,b) Control, (a) Nuclear cell; ( b ) cytoplast; (c,d) after treatment with 0.3 µg/ml nocodazole. (c) Immunofluorescence; (d) phase contrast. Bar = 10 µm.


Resistance of microtubules in cells and cytoplasts to nocodazole treatment

The microtubule system, revealed using immunofluores-cence staining with tubulin antibodies in cytoplasts 2-3 h and 14-16 h after enucleation, resembled as a whole the normal one in the cells (Fig. 2a,b). However, after incubation with nocodazole (0.3 (µg/ml, 1 h), distinctions between cells and cytoplasts became evident. In nocodazole-treated cells only few short microtubules, radiating from the cell center, and few free microtubules could be seen in the cytoplasm. At the same time in most cytoplasts extended microtubules radiating from one center were presented in a greater number (Fig. 2c,d). This difference might be explained in two ways. First, there might be much more microtubules attached in the cen-trosome of cytoplasts compared to the nuclear cells. An-

other explanation is that microtubules in cytoplasts are more stable (less sensitive to nocodazole-induced depoly-merization). To gain insight into this problem we treated cytoplasts and nuclear cells with nocodazole for different times at a concentration of 5 µg/ml. 5 µg/ml of nocodazole induced depolymerization of mi-crotubules in cells and gradual depolymerization in cy-toplasts. After 2 h treatment of cells microtubules were completely depolymerized. The same treatment resulted in the absence of microtubules in cytoplasts within 4 h. After 30 min treatment microtubule patterns in cytoplasts and cells were already different. After 60-90 min treatment long microtubules remained in the cytoplasts (Fig. 3 a). At the same time few microtubules were observed in cells (Fig. 3 b). After 90 min treatment in the majority of cells there were no microtubules. But all cytoplasts contained a number of microtubules up to 3 h of treatment

FIGURE 3a-d Antitubulin staining of cells (b,d) and cytoplasts (a,c) after treatment with 5 µg/ml nocodazole. (a,b) 1 h; (c,d) 3 h. Bar = 10 µm.


FIGURE 4a,b Serial sections of the centrosome in normal cytoplasts (16 h after enucleation). (a) Active centriole; (b) nonactive centriole. LD: lipid . droplets, secondary lysosomes are pointed by arrowheads. Bar = 0.5 µm.


(Fig. 3c). Two patterns of microtubules were evident in the cytoplasts treated with nocodazole for 1-3 h. Typically cytoplasts had nocodazole resistant microtubules radiating from the center (Fig. 3c). However, some cytoplasts had microtubules localized at the periphery of cell body (data not shown). To elucidate whether microtubules resistance to nocodazole in cytoplasts depends on the age of the latter, we made a parallel experiment with cytoplasts 2 h and 16 h after enucleation. The results obtained did not show significant difference between the two. Microtubules resistance to nocodazole in aged cytoplasts was the same as in those just after enucleation (data not shown).

Fine structure of the centrosome

Electron microscopy showed that in cytoplasts 14-16 h after enucleation the centrosome was surrounded by secondary lysosomes, containing lamellar bodies, and by large lipid droplets (Fig. 4a,b). Centrioles in the cytoplasts were usually placed together and located at a different angle to each other. The mother (active) centriole had several pericentriolar satellites and appendages. Sometimes one or two striated rootlets run from the cen-trioles in the cytoplasts (Fig. 4a). In some cytoplasts we found replicated centrioles. Procentrioles were prominent, having their length about 0.2 µm. Besides cylinders of the centrioles and heads of the pericentriolar satellites free foci of microtubule convergence were seldom observed in the cytoplasts (Fig. 6a-d). Ouabain treatment did not make any changes in the centrosome ultrastructure in the cytoplasts (result not shown). But ouabain treatment made nearly half (14 of 32 studied, or 44%) of active centrioles perpendicularly oriented to substrate surface, as it happened in normal cells (Alieva and Vorobjev, 1995). This orientation was significantly different from the random one (p<0.01). As noted above, from immunofluorescence data it might by suggested that not all the cytoplasts have a centrosome. This was confirmed by the electron microscopic study of the complete series of semithin sections, running through the whole lysed cytoplasts. Up to 20% of cytoplasts had no centrioles and about 20% of cytoplasts contained only one centriole, either mother or daughter. In the centriole-free cytoplasts no microtubule convergence centers were found under electron microscope (data not shown).

The stereoscopic analysis of microtubules around the centrosome

The electron microscopic analysis showed that on the av-erage 28.6 ± 4.8 microtubules radiated from the centro-

some in cells; 15.4 ±3 .2 of them were fastened by their proximal ends to one of the two types of nucleating structures: to pericentriolar satellites and to walls of cyl-inders of mother or daughter centrioles (Fig. 5a-с). Some of the microtubules (13.2 ± 3.8) were not associated with any electron-dense structure. Such microtubules had a free proximal end, which was directed to one of the foci of microtubule convergence. Besides that there were 14.1 ± 4.0 cytoplasmic microtubules passing by the centrosome. Some of these microtubules crossed the investi-

FlGURE 5a-c Series of semithin sections of the centrosome in PK cell (lysis prior to fixation). Stereopairs. Bar = 1 µm.


gation volume of the cytoplasm, others were terminated inside this volume (Fig. la) . On the average 50.1 ± 4.4 microtubules radiated from the centrosome in cytoplasts. Two third (33.0 ± 3.2) of them were fastened by their proximal ends to one of the three types of nucleating structures: to pericentriolar satellites, to walls of centriole cylinders, or to free foci (Fig. la,b). The free proximal end had one third of microtubules (17.1 ± 2.0). Besides that 30.2 ± 5.6 microtubules passing by the centrosome were on the average in the cytoplasts (Fig. Ib).

DISCUSSION

Despite the data available in literature that centrioles always remain in cytoplasts as a result of standard enuclea-tion process after cytochalasin treatment (Prescott et al., 1972; Goldman et al., 1975; Zorn et al., 1979; Karsenti et al., 1984), in our experiments up to 20% of the cytoplasts lacked both centrioles and a part of cytoplasts contained only one centriole. This argues for the centrioles to remain in some karyoplasts after centrifugation. Thus appearance of centrioles in regenerated karyoplasts described as a de novo formation of centrioles (Zorn et al., 1979) needs to be reexamined. Some cytoplasts obtained from PK cells contained pro-centrioles. Since 13-15 h passed after enucleation, that is more than 1/2 of the cell cycle of PK cells, one should suppose that either the replication occurred in a cell prior to enucleation and after it procentrioles stopped growing, or that the centrioles started replicating in the cytoplast. These two possibilities are now under examination. In normal PK cells only few microtubules radiate from the centrosome (Alieva et al., 1992). Their number increases under the influence of energy transfer inhibitors (FCCP, 2,4-dinitrophenol, sodium azide) and ouabain (Alieva et al., 1992) and is supplemented with appearance of free foci of microtubule convergence. Similar effect was observed upon spreading of fibroblasts on a substrate (Gudima et al., 1986). However, in some cytoplasts obtained from PK cells we have found free foci of micro-tubule convergence without any additional treatment. Furthermore, in cytoplasts the number of microtubules fastened to the microtubule convergence foci is twofold greater than that in cells. It is worth noting that we compared the nuclear cells and cytoplasts from the same coverslips, i.e. those which were subjected to the same treat-ments with cytochalasin D and to centrifugation. Thus, the nucleus removal from cells appears to be sufficient for centrosome activation. In the works of Gudima with co-authors (Gudima et al.,

FIGURE 6a-d Series of semithin sections of the centrosome in the 16 h cytoplast (lysis prior to fixation). Arrowhead points free focus of micro-tubule convergence. Stereopairs. Bar = 1 µm .


FIGURE 1a,b Profiles of microtubules around the centrosome. (a) In the PK cell (reconstructed from 5 serial sections); (b) in the cytoplast (reconstructed from 6 serial sections). Solid lines: microtubules fastened to the centers; dashed lines: free microtubules.

1986), and Albrecht-Buehler and Bushnell (1979) the nonrandom (preferentially perpendicular to the substrate plane) orientation of active centrioles is assumed to be a result of the centrosome activation, that is the increase in the amount of radiating microtubules. However, the portion of the centrioles, located at an angle over 74° (i.e. perpendicular to substrate plane) after FCCP treatment was shown to increase when microtubules were depoly-merized with nocodazole pretreatment (Alieva and Vo-robjev, 1990). Thus two phenomena appear not to be associated with each other as suggested by Albrecht-Buehler and Bushnell (1979). More than that, it is evident from our data that when the number of centrosome associated microtubules increased as a result of enucleation the distribution of active centrioles in the cytoplasts remained close to accidental. Thus, taking into account previously obtained results, it may be stated that the perpendicular orientation of the centrioles to the substrate plane and the centrosome activation in the same cells are controlled in different ways.

To study microtubules stability we used nocodazole. No-codazole, at a concentration of 0.3 µg/ml, within 1 h de-polymerized mainly nonattached (free) microtubules (Karsenti et al., 1984; Tassin et al., 1985). The results ob-tained demonstrate that only few long microtubules in PK cells are fastened to the centrosome and more microtubules in the cytoplasts. This was confirmed by electron microscopic reconstruction (see above). In the cytoplasts microtubules are slightly depolymerized by low concentrations of nocodazole, and are gradually depolymerized by high concentrations of the drug. It should be stressed that microtubules not converging towards the centrosome in the cytoplasts were also more stable to nocodazole action. That means that all microtubules become nocodazole resistant after enucleation. The existence of microtubule regulating factors in the cytoplasm including Ca2+ and p34ctk2 protein kinase was demonstrated recently (Lieuvin et al., 1994). It was reported that in the absence of these factors MTs become more stable (Lieuvin et al., 1994). Our data give further evidence for microtubule instability to be modulated by regulatory effectors. Effectors, whose action was observed in our experiments, are under cell nucleus control. p34cdc2 protein kinase is a good candidate for this role, but the question needs further investigation. Studying microtubule pattern around the centrosome in the cytoplasts we found an increase in number of microtubules around it and claimed this as centrosome activation. But taking into account higher resistance of microtubules in the cytoplast to nocodazole, it might be suggested that after cell enucleation the main change of the centrosome is not an increase in its microtubule nucleating activity, but slowdown of the microtubule turnover and particularly of releasing of microtubules from the nucleating sites.

ACKNOWLEDGEMENTS

This work was supported in part by grant from Russian Basic Science Foundation to I.A.V.

REFERENCES

ALBRECHT-BUEHLER G. and BUSHNELL A., 1979. The orientation of centrioles in migrating 3T3 cells. Exp. Cell Res., 120, 111-118.

ALIEVA I.B. and VOROBJEV LA., 1989. Centrosome reaction on the action of Ca2"1" ionophore A23187. Tsitologiya (Russian), 31, 259-266.

ALIEVA I.B. and VOROBJEV LA., 1990. The effect of the disruption of cytoskeleton elements on the uncoupler-induced alterations of the centrosome. Tsitologiya (Russian), 32, 620-625.

ALIEVA I.B. and VOROBJEV LA., 1995. Centrosome behaviour and orientation of centrioles under the action of energy transfer inhibitors. Cell Biol. Intern., 19, 103-112.


ALIEVA I.B., VAISBERG E.A., NADEZHDINA E.S. and VOROBJEV I.A., 1992. Microtubule and intermediate filament patterns around the cen-trosome in interphase cells. In: 'The centrosome'. Kalnins V.C. ed., Academic Press, New York, London, Sydney, Toronto, pp. 103-129.

ALVEY P.L., 1986. Do adult centrioles contain cartwheels and lie at right angle to each other? Cell BioL Intern. Rep., 10, 589-599.

BERSHADSKY A.D. and GELFAND V.I., 1981. ATP-dependent regulation of cytoplasmic microtubule assembly. Proc. Natl. Acad. Sci. USA, 78, 3610-3613.

BROWN R.L., WIBLE L.J. and BRINKLEY B.R., 1980. Cytoplasmic assembly-disassembly in enucleated cells and regenerating karyoplasts. Cell Biol. Int. Rep., 4, 453-458.

GOLDMAN R.D., POLLACK R., CHANG S.M. and BUSHNELL A., 1975. Properties of enucleated cells. III. Changes in cytoplasmic architecture of enucleated BHK-21 cells following trypsinization and replat-ing. Exp. Cell Res., 93, 175-183.

GUDIMA G.O., VOROBJEV LA. and CHENTSOV Yu.S., 1986. Cell center behavior during fibroblast spreading. 1. Normal fibroblasts. Proc. Xlth Int. Cong. on Electron Microscopy, Kyoto, pp. 2523-2524.

JOHNSON L.V., WALSH M.L. and CHEN L.B., 1980. Localization of mi-tochondria in living cells with rhodamine 123. Proc. Natl. Acad. Sci. USA, 77, 990-994.

JOHNSON L.V., WALSH M.L., BOOKUS B.J. and CHEN L.B., 1981. Monitoring of relative mitochondria! membrane potential in living cells by fluorescence microscopy. /. Cell Biol., 88, 526-534.

JOSHI H.C. and BAAS P.W., 1993. A new perspective on microtubules and axon growth. /. Cell Biol., 121, 1191-1196.

KARSENTI E., KOBAYASHI S., MITCHISON T. and KIRSCHNER M., 1984. The role of centrosome in organizing of microtubule array: proper-

ties of cytoplasts containing or lacking centrosomes. /. Cell BioL, 98, 1763-1776.

KURIYAMA R. and BORISY G.G., 1981. Microtubule-nucleating activity of centrosomes in Chinese hamster ovary cells is independent of the centriole cycle but coupled to the mitotic cycle. /. Cell BioL, 91, 822-826.

LIEUVIN A., LABBE J.-C., DOREE M. and JOB D., 1994. Intrinsic microtubule stability in interphase cells. /. Cell Biol., 124, 985-996.

PETERSON S.P. and BERNS M.W., 1980. The centriolar complex. Int. Rev. CytoL, 64, 81-106.

PRESCOTT D.M., MYERSON D. and WALLANSE G., 1972. Enucleation of mammalian cells with cytochalasin B. Exp. Cell Res., 71, 480-485.

SCHLIWA M., EUTENEUER U., HERZOG W. and WEBER K., 1979. Evidence for rapid structural and functional changes of the melan-ophore microtubule-organizing center upon pigment movement. /. Cell Biol., 83, 623-632.

TASSIN A.-M, MARO B. and BORNENS M., 1985. Fate of microtubule-organizing centers during myogenesis in vitro. ]. Cell Biol., 100, 35-46.

VASILYEV I.A., VOROBJEV I.A., LEONTOVICH A.M. and PETROVSKAYA M.B., 1988. The analysis of orientation of centrioles in tissue culture cells. Tsitologiya (Russian), 30, 1091-1100.

VOROBJEV LA. and CHENTSOV Yu.S., 1982. Centrioles in the cell cycle. I. Epithelial cells. /. Cell BioL, 98, 938-949.

VOROBJEV I.A. and NADEZHDINA E.S., 1987. The centrosome and its role in the organization of microtubules. Int. Rev. CytoL, 106, 227-293.

ZORN G.A., LUCAS J.J. and KATES J.R., 1979. Purification and charac-terization of regenerating mouse L929 karyoplasts. Cell, 18, 659-672.


centrosome and microtubules behavior in the...

Documents